ACOX2

Summary: The ancestral fatty-acid-binding assertion needs evaluation separately from acyl-CoA substrate recognition.

Reason: The recorded PAINT lineage places the fatty-acid-binding IBD at PTN000097533 on the Q99424 path. Rat Acox2 and Acox3 are the experimental descendants; source tracing identifies older rat substrate studies, but their full free-acid-binding evidence could not be adjudicated. GO:0005504 denotes fatty-acid binding, not automatically recognition of a CoA ester. Neither preferred ester substrates nor incomplete donor access establishes loss of the ancestral function. Retain UNDECIDED without rejecting PAINT on donor count or assuming that acyl-CoA oxidases cannot bind free acids. Independent family-level context in PMID:16672280 is positive: rat ACOX1 (ACO-II; structure 2DDH, UniProt P07872) contains a free C12 fatty acid in its active-site crevice after hydrolysis of the cocrystallized dodecanoyl-CoA. This establishes accommodation of a C12-derived acid after thioester cleavage, not a binding assay using an externally supplied free fatty acid. It is family context from ACOX1, not the uninspected Acox2/Acox3 descendant experiment underlying this IBD, and does not resolve transfer to human ACOX2.

Summary: Phylogenetically inferred involvement in fatty acid beta-oxidation via an acyl-CoA oxidase. This is exactly the pathway ACOX2 participates in, catalyzing the first oxidase step of peroxisomal beta-oxidation of branched-chain and bile-acid CoA esters.

Reason: ACOX2 catalyzes an oxidation step in peroxisomal beta-oxidation. The conserved PAINT process agrees with the human recombinant substrate assays in PMID:29287774, the bile-acid pathway phenotype in PMID:27884763, and the cached UniProt pathway assignment. The human whole-cell assay establishes pathway involvement; it is not an isolated elementary-reaction assay.

Summary: Human ACOX2 has weak but measurable straight-chain acyl-CoA oxidase activity.

Reason: Full PMID:29287774 directly measures human ACOX2 activity toward C10-CoA and C16-CoA in a yeast fox1 deletion background lacking endogenous acyl-CoA oxidase. Low activity compared with pristanoyl-CoA does not negate the reaction. Retain medium/long-chain oxidation as secondary capacity; it does not establish predominant physiological flux or activity toward very-long-chain C24/C26 substrates.

Summary: Phylogenetically inferred peroxisomal localization/activity. ACOX2 carries a C-terminal SKL (PTS1) signal and is directly demonstrated to be peroxisomal in human liver, and is active within the peroxisome.

Reason: The ancestral peroxisomal localization agrees with human localization evidence in PMID:8943006 and PMID:27884763, the C-terminal SKL signal, and the UniProt subcellular-location record. Retain this supported organelle-level annotation at its source resolution.

Summary: FAD binding inferred phylogenetically. ACOX2 is a FAD-dependent flavoprotein oxidase; FAD is an essential cofactor for the oxidative dehydrogenation reaction.

Reason: FAD is the redox cofactor recorded in UniProt and the ACOX2:FAD Reactome reactions. This is consistent with the ancestral PAINT cofactor-binding assignment and the FAD-supplemented human oxidase assays in PMID:29287774. The UniProt cofactor statement is itself by similarity; it is not presented as an independent human binding measurement.

Summary: The very small C24-CoA signal in a partially purified human preparation does not securely resolve ACOX2-specific VLCFA activity.

Reason: PMID:8387517 Table III reports lignoceroyl-CoA turnover of 0.14 versus 59.4 nmol H2O2/min/ml for 2-methylpalmitoyl-CoA (0.2%) in a chromatographically separated, partially purified preparation. This establishes a preparation-level signal, but does not securely distinguish intrinsic ACOX2 turnover from residual activity of another enzyme. Contamination is a possibility, not a demonstrated fact. The recombinant human panel in PMID:29287774 tested C10/C16, not C24/C26, so it does not resolve this attribution. Keep the inherited VLCFA-process annotation UNDECIDED; normal patient VLCFA concentrations and total C27 steroid carbon count do not settle that question. The Table III caption on page 10343 was read in the publicly accessible author-uploaded original, linked in the notes' Public primary-text receipts; that full-text passage is absent from the normal abstract-only cache.

Summary: THCA-CoA oxidase evidence is represented with the verified oxidase activity pending correction of the hydroxylase term.

Reason: The cached PAINT IBD at PTN008508564 contains human and rat descendants. Human target evidence is legitimate grounding, not circular. The exact rat assay remains unresolved, but the independently verified human reaction and current term definition establish the target-level chemistry mismatch. GO:0033791 currently defines a 25R-substrate hydroxylation using water and an acceptor, whereas the supported ACOX2 reaction is oxygen-dependent oxidation of a 25S substrate to a 24E-enoyl product (UniProt RHEA:46728). The hydroxylase term has historical oxidase synonyms, but its formal reaction is different. Use GO:0003997 for the supported oxidase chemistry pending ontology correction, preserving THCA-CoA specificity in the evidence and core description. This replacement duplicates an already supported broad activity rather than creating new functional coverage; it does not assert universal absence of coupled hydroxylation. Rat-liver oxidases in PMID:8654595 directly show S-selectivity with 2-methylpentadecanoyl-CoA. Its abstract reports desaturation of one THCA-CoA epimer without identifying that epimer or the product geometry; the exact 25S-to-24E reaction above is the curated human UniProt assignment, not a claim that the rat abstract is a stereochemical assay of human ACOX2.

Proposed replacements: acyl-CoA oxidase activity

Summary: Acyl-CoA oxidase activity (EC 1.3.3.6) assigned electronically from InterPro/Rhea. This is the defining molecular function of ACOX2 and is experimentally confirmed.

Reason: The generic oxidase reaction matches the cached UniProt RHEA:38959/EC:1.3.3.6 assignment and human substrate experiments in PMID:29287774. PMID:27884763 independently supports the bile-acid pathway through whole-cell conversion; it does not by itself measure the isolated oxidase stoichiometry. The broad molecular function is accurate for both bile-acid and branched-chain acyl-CoA substrates.

Summary: Peroxisomal localization assigned electronically (InterPro / UniProt SubCell). Confirmed by direct experimental evidence.

Reason: Consistent with the UniProt SUBCELLULAR LOCATION (Peroxisome) and multiple experimental annotations (IDA/IMP). The SKL C-terminal PTS1 targets the protein to the peroxisome.

Summary: Peroxisomal matrix localization assigned by ARBA machine-learning model. As a soluble PTS1-targeted matrix enzyme, ACOX2 resides in the peroxisomal matrix, consistent with Reactome placing its reactions in the peroxisomal matrix.

Reason: The matrix assignment agrees with PTS1-directed localization and the compartment of the curated ACOX2 reactions in Reactome:R-HSA-192335 and Reactome:R-HSA-193369. The ARBA rule predicates were not independently inspected; acceptance of the location rests on this independent biological evidence.

Summary: Refine the broad fatty-acid metabolism annotation to the demonstrated oxidase-dependent beta-oxidation pathway.

Reason: Human recombinant substrate assays establish ACOX2 acyl-CoA oxidation, and the whole-cell bile-acid conversion assay confirms its contribution to peroxisomal side-chain shortening. These independent data support the specific beta-oxidation process rather than retaining only the broad InterPro-derived metabolic category.

Proposed replacements: fatty acid beta-oxidation using acyl-CoA oxidase

Summary: Fatty acid beta-oxidation assigned electronically from InterPro. ACOX2 catalyzes the first oxidase step of peroxisomal beta-oxidation, so this is an accurate parent of the more specific "fatty acid beta-oxidation using acyl-CoA oxidase" term.

Reason: Directly supported by the UniProt PATHWAY (peroxisomal fatty acid beta-oxidation) and by experimental data. This is a core biological process for ACOX2.

Summary: Pristanoyl-CoA oxidase activity assigned from Rhea (RHEA:40459). ACOX2 oxidizes (2S)-pristanoyl-CoA, a 2-methyl-branched-chain fatty acyl-CoA, though it acts redundantly with ACOX3 for this substrate.

Reason: Supported by the UniProt catalytic activity for (2S)-pristanoyl-CoA (Rhea:40459, ECO:0000269|PubMed:29287774). This is a genuine branched-chain fatty acyl-CoA oxidase activity of ACOX2.

Summary: Refine the broad oxidoreductase annotation to acyl-CoA oxidase activity.

Reason: Human substrate assays and the UniProt catalytic reaction specify acyl-CoA oxidation with oxygen reduction and hydrogen-peroxide formation. They support the specific molecular function. The ARBA rule predicates were not inspected independently; the refinement follows the independent chemistry.

Proposed replacements: acyl-CoA oxidase activity

Summary: Refine the CH-CH oxidoreductase class to the established acyl-CoA oxidase activity.

Reason: The InterPro mapping is chemically correct but lacks the acyl-CoA substrate and oxygen acceptor specificity. Human recombinant assays and the curated catalytic reaction support the more informative oxidase term.

Proposed replacements: acyl-CoA oxidase activity

Summary: Fatty acid beta-oxidation using acyl-CoA oxidase, assigned by ARBA. This is the precise pathway term for ACOX2 and is also supported experimentally.

Reason: ACOX2 performs the oxidase step of peroxisomal beta-oxidation, as supported by human substrate studies and the cached UniProt pathway record. The ARBA rule predicates were not independently inspected. Acceptance rests on this biological evidence, not merely on other rows carrying the same term.

Summary: InterPro-derived flavin adenine dinucleotide binding is consistent with the oxidase cofactor.

Reason: The cached UniProt cofactor and FAD-containing Reactome enzyme support flavin adenine dinucleotide binding. The InterPro signature provides family evidence; it is not an independent measurement of binding affinity.

Summary: GO:0071949 specifies binding of oxidized FAD.

Reason: The live AmiGO definition specifies oxidized FAD and places this term under GO:0050660, which covers flavin adenine dinucleotide binding more generally. These terms are not identical. The oxidized cofactor is appropriate for the substrate-oxidizing half-reaction and is explicitly recorded as FAD in UniProt; retain the source annotation.

Summary: Human ACOX2 has secondary C10-CoA oxidase activity.

Reason: The C10-CoA reaction in the cached UniProt Rhea entry agrees with the recombinant human activity measured in PMID:29287774. Retain this weaker straight-chain capacity as non-core relative to the branched-chain and bile-acid functions. This source-specific substrate assignment does not establish predominant physiological flux.

Summary: Human ACOX2 has secondary C16-CoA oxidase activity.

Reason: The C16-CoA reaction in the cached UniProt Rhea entry agrees with the recombinant human activity measured in PMID:29287774. Retain this weaker straight-chain capacity as non-core relative to the branched-chain and bile-acid functions. This source-specific substrate assignment does not establish predominant physiological flux.

Summary: The BioPlex 2.0 affinity-purification mass spectrometry annotation records an association with STRN3 (Q13033).

Reason: GOA and the cached UniProt interaction record identify STRN3 (Q13033). Generic protein binding adds no specific molecular function, and the available study context does not establish an evidence-backed replacement for ACOX2. REMOVE concerns the uninformative functional annotation, not a claim that the association is false. The individual supplementary pair record was not independently revalidated.

Summary: The HuRI binary-interaction mapping annotation records an association with DYNLT1 (P63172).

Reason: GOA and the cached UniProt interaction record identify DYNLT1 (P63172). Generic protein binding adds no specific molecular function, and the available study context does not establish an evidence-backed replacement for ACOX2. REMOVE concerns the uninformative functional annotation, not a claim that the association is false. The individual supplementary pair record was not independently revalidated.

Summary: The BioPlex 3.0 affinity-purification mass spectrometry annotation records an association with STRN3 (Q13033).

Reason: GOA and the cached UniProt interaction record identify STRN3 (Q13033). Generic protein binding adds no specific molecular function, and the available study context does not establish an evidence-backed replacement for ACOX2. REMOVE concerns the uninformative functional annotation, not a claim that the association is false. The individual supplementary pair record was not independently revalidated. The publication cache is incomplete despite its full-text metadata flag.

Summary: The R225W family has accumulated C27 intermediates and depleted C24 bile acids. Wild-type ACOX2 expression enhances THCA-to-cholic-acid conversion in HepG2 cells, whereas the R225W protein does not. This supports the enzyme contribution to bile-acid biosynthesis.

Reason: The source demonstrates increased THCA-to-cholic-acid conversion after wild-type ACOX2 expression in HepG2 cells, with impaired conversion for R225W. This supports a catalytic contribution to bile-acid biosynthesis. The affected siblings' abnormal C27/C24 profiles corroborate the pathway role; negligible C24 bile acids are a finding in this family, not a universal definition of ACOX2 deficiency. Independently, PMID:27647924 reports a homozygous p.Y69* premature termination variant, absent ACOX2 staining in patient liver, and accumulated bile-acid intermediates upstream of ACOX2. This human deficiency phenotype corroborates the pathway contribution; it is not a separate isolated-enzyme reaction assay.

Summary: ACOX2 can occupy the cytosol as a PEX5 cargo before peroxisomal import.

Reason: Reactome:R-HSA-9033235 models PEX5-bound cargo before translocation and release into the peroxisomal matrix. The cytosolic location is appropriate for this import intermediate and is retained as non-core. It does not establish a cytosolic catalytic function.

Summary: ACOX2 can occupy the cytosol as a PEX5 cargo before peroxisomal import.

Reason: Reactome:R-HSA-9033236 models docking of PEX5-bound PTS1 cargo at the translocation module. The source annotation represents the cytosolic import precursor and is retained as non-core. ACOX2 is the cargo in this event; no import-machinery function is inferred.

Summary: Peroxisomal localization confirmed by immunofluorescence in HepG2 cells overexpressing wild-type and R225W-mutant ACOX2; both showed peroxisomal localization. Direct experimental support for the peroxisome location.

Reason: PMID:27884763 reports peroxisomal localization of both wild-type and R225W ACOX2 in HepG2 cells. The publisher methods specify colocalization with PMP70. Retain the demonstrated location; the variant's pathway defect is not explained by loss of peroxisomal targeting.

Summary: Direct experimental evidence for ACOX2's role in acyl-CoA-oxidase-mediated beta-oxidation, shown via its bile-acid intermediate (THCA) processing that requires functional ACOX2. Core biological process.

Reason: Wild-type ACOX2 increases THCA-to-cholic-acid conversion in HepG2 cells in PMID:27884763, supporting participation in peroxisomal beta-oxidation of the bile-acid side chain. This whole-cell endpoint does not separately resolve every chemical step. Independent recombinant human oxidase experiments in PMID:29287774 corroborate the enzyme's contribution.

Summary: THCA-CoA oxidase evidence is represented with the verified oxidase activity pending correction of the hydroxylase term.

Reason: The full human source PMID:27884763 measures ACOX2-dependent THCA-to-cholic-acid conversion in HepG2 cells over 48 hours. That result establishes pathway function rather than isolated hydroxylation. The recombinant human substrate study PMID:29287774 and curated reaction record independently support the oxidase assignment. GO:0033791 currently defines a 25R-substrate hydroxylation using water and an acceptor, whereas the supported ACOX2 reaction is oxygen-dependent oxidation of a 25S substrate to a 24E-enoyl product (UniProt RHEA:46728). The hydroxylase term has historical oxidase synonyms, but its formal reaction is different. Use GO:0003997 for the supported oxidase chemistry pending ontology correction, preserving THCA-CoA specificity in the evidence and core description. This replacement duplicates an already supported broad activity rather than creating new functional coverage; it does not assert universal absence of coupled hydroxylation. Rat-liver oxidases in PMID:8654595 directly show S-selectivity with 2-methylpentadecanoyl-CoA. Its abstract reports desaturation of one THCA-CoA epimer without identifying that epimer or the product geometry; the exact 25S-to-24E reaction above is the curated human UniProt assignment, not a claim that the rat abstract is a stereochemical assay of human ACOX2.

Proposed replacements: acyl-CoA oxidase activity

Summary: Peroxisomal localization transferred by sequence similarity from ACOX1 (UniProtKB:P07872). Consistent with the direct experimental peroxisomal localization of ACOX2.

Reason: The ISS donor P07872 is rat ACOX1, a paralog rather than rat ACOX2. The peroxisomal compartment can be conserved across these family members, and the human location is independently supported by PMID:8943006 and PMID:27884763. Accept the location without presenting this as ortholog-specific evidence.

Summary: R225W fails to restore normal THCA-to-cholic-acid conversion in the tested human-cell system. Together with the affected-family bile-acid profiles, this supports participation in acyl-CoA-oxidase-mediated beta-oxidation.

Reason: In PMID:27884763, R225W fails to enhance THCA-to-cholic-acid conversion despite similar expression and peroxisomal localization. Together with the affected-family bile-acid profiles, this supports ACOX2 involvement in the oxidase-dependent beta-oxidation pathway. It does not imply a defect in every fatty-acid substrate or directly resolve the disputed hydroxylase stoichiometry.

Summary: THCA-CoA oxidase evidence is represented with the verified oxidase activity pending correction of the hydroxylase term.

Reason: The donor O02767 is rabbit ACOX2. Its primary study PMID:9218493 identifies the 24E-enoyl product by chromatography and mass spectrometry; the small hydroxylated product is attributed most probably to endogenous COS-cell hydratase. This is positive support for donor oxidase chemistry, with the hydratase explanation remaining probabilistic. The external original Results, page 18487 beside Figures 9-10, were read at the public PDF linked in the notes' Public primary-text receipts; this passage is absent from the normal abstract-only cache. GO:0033791 currently defines a 25R-substrate hydroxylation using water and an acceptor, whereas the supported ACOX2 reaction is oxygen-dependent oxidation of a 25S substrate to a 24E-enoyl product (UniProt RHEA:46728). The hydroxylase term has historical oxidase synonyms, but its formal reaction is different. Use GO:0003997 for the supported oxidase chemistry pending ontology correction, preserving THCA-CoA specificity in the evidence and core description. This replacement duplicates an already supported broad activity rather than creating new functional coverage; it does not assert universal absence of coupled hydroxylation. Rat-liver oxidases in PMID:8654595 directly show S-selectivity with 2-methylpentadecanoyl-CoA. Its abstract reports desaturation of one THCA-CoA epimer without identifying that epimer or the product geometry; the exact 25S-to-24E reaction above is the curated human UniProt assignment, not a claim that the rat abstract is a stereochemical assay of human ACOX2.

Proposed replacements: acyl-CoA oxidase activity

Summary: The human enzyme's native oligomeric state is not settled by the rat ACOX1 transfer.

Reason: The donor P07872 is rat ACOX1, and UniProt assigns human ACOX2 homodimerization by similarity. PMID:8387517 reports approximately 67-70 kDa human preparations but contrasts an earlier approximately 138 kDa estimate and explicitly leaves native assembly unresolved, including possible dissociation during purification. Reactome:R-HSA-389889 retains the monomeric description. These data neither establish constitutive human homodimerization nor exclude it; retain UNDECIDED rather than generalize across every acyl-CoA oxidase. The native-mass qualification in the external original Discussion, page 10343, was read in the author-uploaded article linked in the notes' Public primary-text receipts; it is absent from the normal abstract-only cache.

Summary: FAD binding transferred by sequence similarity from ACOX1 (UniProtKB:P07872). ACOX2 is a FAD flavoprotein. Redundant with the IBA/IEA FAD-binding annotations.

Reason: The donor P07872 is rat ACOX1. FAD-dependent oxidase chemistry is shared with ACOX2, and the cached UniProt cofactor statement explicitly records this by-similarity transfer. Human ACOX2 reactions and FAD-supplemented assays are compatible with it; no paralog-specific loss of cofactor use is indicated.

Summary: Peroxisomal matrix localization from Reactome, tied to the curated reaction in which ACOX2 dehydrogenates 25(S) THCA-CoA in the peroxisomal matrix. Accurate compartment for this soluble matrix enzyme.

Reason: Reactome explicitly places the ACOX2-catalyzed THCA-CoA dehydrogenation in the peroxisomal matrix. This is the accurate, specific compartment for the soluble PTS1-targeted enzyme.

Summary: Peroxisomal matrix localization from Reactome, tied to the curated reaction in which ACOX2 dehydrogenates 25(S) DHCA-CoA. Accurate compartment for this soluble matrix enzyme.

Reason: Reactome places the ACOX2-catalyzed DHCA-CoA dehydrogenation in the peroxisomal matrix, consistent with the enzyme's soluble matrix localization. Accurate, specific compartment.

Summary: Peroxisomal matrix localization from Reactome, tied to the curated reaction in which ACOX2:FAD oxidizes (2S)-pristanoyl-CoA. Accurate compartment for this soluble matrix enzyme.

Reason: The curated reaction assigns the peroxisomal ACOX2:FAD catalyst to the matrix, consistent with the other reaction compartments and PTS1 localization. Its summary's monomeric description is a separate assembly claim; it does not invalidate the location and is not used to establish native stoichiometry.

Summary: Peroxisomal matrix localization from Reactome, from the peroxisomal-import reaction in which PEX5 cargo (including ACOX2) is translocated into the matrix. Accurate destination compartment.

Reason: The reaction describes cargo translocation into the peroxisomal matrix, which is ACOX2's functional destination. Consistent with the other matrix annotations and experimental data.

Summary: Direct experimental peroxisomal localization: the human branched-chain acyl-CoA oxidase carries a C-terminal SKL PTS1 signal and was shown to be peroxisomal (and absent from Zellweger livers by immunoblot/immunocytochemistry). Core localization.

Reason: The cached abstract of PMID:8943006 identifies the cloned human protein and its C-terminal SKL targeting signal and reports immunocytochemistry in control and Zellweger liver. Full figure-level localization was not recovered in this session. Retain the curator's peroxisomal IDA, independently corroborated by the human localization study PMID:27884763; absence in Zellweger tissue alone is not treated as a complete localization assay.

Summary: Human-liver enzymology identified a distinct peroxisomal trihydroxycoprostanoyl-CoA oxidase before molecular cloning.

Reason: The accessible abstract of PMID:2079609 reports separated human-liver oxidase activities and a distinct trihydroxycoprostanoyl-CoA oxidase. This predates sequence assignment to ACOX2. Retain the curator's peroxisomal localization, consistent with later molecular and cell-localization evidence, without claiming that the 1990 study cloned or sequence-identified the gene.